PO87

Abstract

Purpose To propose a generalization of the known relationship between total reference air Kerma (TRAK) and isodose surface volumes for intracavitary, hybrid and interstitial applicators used for treating cervical cancer with high dose rate (HDR) brachytherapy (BT). Materials and Methods A single institution cohort of 123 retrospective clinical HDR BT plans from 34 patients treated for cervical cancer were evaluated. The cohort consisted of 71 intracavitary (tandem and ring - T&R - and tandem and ovoid - T&O), 32 hybrid (T&R or T&O with the addition of stainless steel and/or flexi needles) and 20 interstitial plans. Patients received 45Gy external beam radiotherapy (EBRT), followed by one of three fractionation schemes delivered with BT: 600cGy x 4 fractions (4 patients for a total of 16 fractions), 700cGy x 4 fractions (21 patients for a total of 80 fractions) and 800cGy x 3 fractions (9 patients for a total of 27 fractions). The average dose per fraction was 708.9±58.5cGy considering all 123 plans. For each plan the isodose surface volumes (TPSvol) were evaluated considering the accumulated EBRT and BT dose. Because three different fractionation schemes were used, the radiobiological equivalent doses in 2Gy fractions (EQD2) were estimated considering the EBRT and BT contributions. We have considered α/β ratio = 10Gy for tumor repair and repair half time T1/2 = 1.5 hour. In this work we have considered three reference dose levels (dref): 60Gy, 75Gy and 85Gy. Figure 1A-C illustrates the isodose surface volumes for the different fractionation schemes. The TRAK of each plan was also recorded. The relationship between TRAK/dref and TPSvol for the different applicators was evaluated by applying a second degree polynomial linear regression considering the two variables for each case. Results The linear regressions showed correlation coefficients R2 of 0.998, 0.997, 0.995 and 0.997 for the data obtained from treatments using intracavitary (Fig. 1D), hybrid (Fig. 1E), interstitials (Fig. 1F) and all applicators together (Fig. 1G), respectively. The linear regressions were not found to be affected by the different fractionation schemes. The quadratic, linear coefficients and the curve intercepts ranged from 0.621 to 0.739, 11.29 to 12.64 and -16.9 to -12.32, respectively. The fitted equation for the hybrid implants (Fig. 1E) showed the largest differences for the quadratic coefficient and curve intercept when compared to the equation fitted for intracavitary and interstitial applicators. The equation resulting from all applicators (Fig. 1G) showed the smallest differences for quadratic and linear coefficients when compared to the equation resulting intracavitary applicators. Conclusions We have shown that TRAK might be useful to predict volumes of isodose surfaces independently of the applicator and fractionation scheme used for treating cervical cancer with BT. The potential to use the correlation between TRAK and volumes of isodose surfaces to predict patients’ outcomes and toxicities should be evaluated in a further study. To propose a generalization of the known relationship between total reference air Kerma (TRAK) and isodose surface volumes for intracavitary, hybrid and interstitial applicators used for treating cervical cancer with high dose rate (HDR) brachytherapy (BT). A single institution cohort of 123 retrospective clinical HDR BT plans from 34 patients treated for cervical cancer were evaluated. The cohort consisted of 71 intracavitary (tandem and ring - T&R - and tandem and ovoid - T&O), 32 hybrid (T&R or T&O with the addition of stainless steel and/or flexi needles) and 20 interstitial plans. Patients received 45Gy external beam radiotherapy (EBRT), followed by one of three fractionation schemes delivered with BT: 600cGy x 4 fractions (4 patients for a total of 16 fractions), 700cGy x 4 fractions (21 patients for a total of 80 fractions) and 800cGy x 3 fractions (9 patients for a total of 27 fractions). The average dose per fraction was 708.9±58.5cGy considering all 123 plans. For each plan the isodose surface volumes (TPSvol) were evaluated considering the accumulated EBRT and BT dose. Because three different fractionation schemes were used, the radiobiological equivalent doses in 2Gy fractions (EQD2) were estimated considering the EBRT and BT contributions. We have considered α/β ratio = 10Gy for tumor repair and repair half time T1/2 = 1.5 hour. In this work we have considered three reference dose levels (dref): 60Gy, 75Gy and 85Gy. Figure 1A-C illustrates the isodose surface volumes for the different fractionation schemes. The TRAK of each plan was also recorded. The relationship between TRAK/dref and TPSvol for the different applicators was evaluated by applying a second degree polynomial linear regression considering the two variables for each case. The linear regressions showed correlation coefficients R2 of 0.998, 0.997, 0.995 and 0.997 for the data obtained from treatments using intracavitary (Fig. 1D), hybrid (Fig. 1E), interstitials (Fig. 1F) and all applicators together (Fig. 1G), respectively. The linear regressions were not found to be affected by the different fractionation schemes. The quadratic, linear coefficients and the curve intercepts ranged from 0.621 to 0.739, 11.29 to 12.64 and -16.9 to -12.32, respectively. The fitted equation for the hybrid implants (Fig. 1E) showed the largest differences for the quadratic coefficient and curve intercept when compared to the equation fitted for intracavitary and interstitial applicators. The equation resulting from all applicators (Fig. 1G) showed the smallest differences for quadratic and linear coefficients when compared to the equation resulting intracavitary applicators. We have shown that TRAK might be useful to predict volumes of isodose surfaces independently of the applicator and fractionation scheme used for treating cervical cancer with BT. The potential to use the correlation between TRAK and volumes of isodose surfaces to predict patients’ outcomes and toxicities should be evaluated in a further study.